To investigate GABA levels in anterior cingulate cortex (ACC) and occipital cortex (OC) in major depressive disorder (MDD) and primary insomnia with proton magnetic resonance spectroscopy imaging (1H- MRS). BACKGROUND: The prevalence of MDD as well as its personal and societal costs have motivated over four decades of research into its pathophysiology and neurobiological correlates. Although the preponderance of studies have focused on the role of the biogenic amines and the hypothalamic-pituitary-adrenal (HPA) axis, recent evidence suggests that the neurotransmitter gamma-aminobutyric acid (GABA) may play a key role in the pathophysiology of MDD. Supporting this conclusion are recent studies utilizing proton magnetic resonance spectroscopy (1H-MRS), which permits in vivo analysis of brain neurochemistry. Most of these studies concurred in finding reduced GABA levels in multiple brain regions in MDD including the OC and ACC. Recently, we utilized 1H-MRS to reveal GABA deficits in subjects with primary insomnia, e.g., insomnia without other accompanying medical or psychiatric disorders. Relative to good sleeper controls, subjects with primary insomnia not only had reduced levels of GABA, but within subjects with primary insomnia, GABA levels correlated with polysomnographic (PSG) measures of sleep maintenance. The proposed study will test the hypothesis that GABA deficits in MDD in fact reflect the sleep disturbance characteristic of that disorder rather than the mood state per se. Confirmation of this hypothesis could have a significant impact on management strategies and guidelines for the pharmacological treatment of this subgroup (those with comorbid MDD and insomnia), which constitutes the majority of patients with MDD. Direct treatment of comorbid insomnia in these patients may contribute to a positive clinical outcome through the restorative effects of sleep, as has been suggested by other clinical studies. Left untreated, insomnia may have a negative impact on clinical outcome and prospects for relapse. METHOD: We plan to perform 1H-MRS imaging for GABA levels (referenced to both creatine and water) in the ACC and OC in 50 subjects with MDD (half of whom will have insomnia and the other half will not), 25 subjects with primary insomnia (insomnia without mood disturbance) and 25 control subjects. All subjects will have sleep diaries and actigraphy for 2 weeks prior to their PSG and MRI. The primary endpoint, GABA levels, will be analyzed by repeated measures ANOVA (using linear mixed model) with diagnostic group (4 study groups) as the main effect and the 2 ROIs (OC and ACC) as repeats. We will also use regression analysis to correlate PSG and sleep diary information, with 1H-MRS-derived GABA data.